EngineersofAI LogoCoursesBlogResearch LabAI LettersThe LabCode BankInteractive 3DKodrEarnestJobs
Skip to main content
Interactive 3D/MLflow Experiment Tracking
MLflow Experiment Runs
Compare hyperparameter configurations across runs. Click a row to inspect its training curve.
Run ID
LR
Batch
Dropout
Val Acc
Val Loss
F1
Status
run_d1e9BEST
0.001
32
0.44
0.8794
0.2525
0.8765
FINISHED
run_g2b3
0.001
32
0.48
0.8028
0.3176
0.7763
FINISHED
run_b7c1
0.001
32
0.42
0.8013
0.3189
0.7641
FINISHED
run_a3f2
0.001
128
0.39
0.7727
0.3432
0.7539
FINISHED
run_f5a8
0.001
32
0.32
0.5302
0.5493
0.5281
FAILED
Training Curve - run_a3f2(lr=0.001, batch=128)
Epoch 1Final val_accuracy: 0.7727Epoch 20
Sort Runs By
Simulate New Run
Add a new experiment run with random hyperparameters.
Legend
Best run (top metric)
Selected run
Failed run
Key Insight
MLflow logs params, metrics, and artifacts per run. The registry promotes the best model to production after comparison.

MLflow Experiment Tracking - Interactive Visualization

MLflow is the de-facto standard for ML experiment tracking. Every training run logs its hyperparameters (learning rate, batch size, dropout), metrics (val_accuracy, val_loss, F1), and artifacts. The experiment UI lets you sort runs by any metric, compare training curves, and identify the best configuration. The best model is then promoted to the MLflow Model Registry for deployment.

  • Compare runs by val_accuracy, val_loss, or F1 - click column headers to sort
  • Inspect training curves for any run - see how accuracy improves over epochs
  • Best run is highlighted in green - MLflow automatically tracks the champion model
  • Simulate new runs with randomized hyperparameters to see how the leaderboard changes

Part of the EngineersOfAI Interactive 3D - free interactive visualizations covering every major concept in machine learning and AI engineering. Hover any element for a plain-English explanation. No code required.